import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sn
pd.set_option('display.max_columns', None)

df = pd.read_csv('../data/clean/stats_salaries.csv')
df.head()

	PLAYER_ID	PLAYER_NAME	START_POSITION	MIN	FGM	FGA	FG_PCT	FG3M	FG3A	FG3_PCT	FTM	FTA	FT_PCT	OREB	DREB	REB	AST	STL	BLK	TO	PF	PTS	PLUS_MINUS	TEAM_ABBREVIATION	TEAM_NAME	GAME_DATE_EST	SEASON_START	WON_GAME	SALARY	INFLATION_ADJ_SALARY	TEAM_PAYROLL	INFLATION_ADJ_TEAM_PAYROLL	LEAGUE_PAYROLL	INFLATION_ADJ_LEAGUE_PAYROLL	TEAM_IMPORTANCE	LEAGUE_IMPORTANCE	TEAM_MARKET_SIZE
0	1630162	Anthony Edwards	F	36:22	4.0	10.0	0.400	3.0	8.0	0.375	4.0	4.0	1.000	0.0	8.0	8.0	5.0	3.0	1.0	1.0	1.0	15.0	5.0	MIN	Minnesota Timberwolves	2022-03-12	2021	1	10245480.0	10245480	137098327	137098327	4125163242	4125163242	0.074731	0.002484	0.033235
1	1630162	Anthony Edwards	F	34:27	9.0	19.0	0.474	4.0	11.0	0.364	3.0	3.0	1.000	0.0	3.0	3.0	5.0	1.0	0.0	0.0	3.0	25.0	-3.0	MIN	Minnesota Timberwolves	2022-03-11	2021	0	10245480.0	10245480	137098327	137098327	4125163242	4125163242	0.074731	0.002484	0.033235
2	1630162	Anthony Edwards	F	25:29	7.0	15.0	0.467	2.0	8.0	0.250	0.0	0.0	0.000	0.0	1.0	1.0	3.0	2.0	1.0	4.0	2.0	16.0	7.0	MIN	Minnesota Timberwolves	2022-03-09	2021	1	10245480.0	10245480	137098327	137098327	4125163242	4125163242	0.074731	0.002484	0.033235
3	1630162	Anthony Edwards	F	32:22	7.0	13.0	0.538	1.0	5.0	0.200	2.0	2.0	1.000	2.0	2.0	4.0	4.0	4.0	0.0	4.0	5.0	17.0	7.0	MIN	Minnesota Timberwolves	2022-02-28	2021	1	10245480.0	10245480	137098327	137098327	4125163242	4125163242	0.074731	0.002484	0.033235
4	1630162	Anthony Edwards	F	37:46	5.0	13.0	0.385	1.0	6.0	0.167	4.0	6.0	0.667	1.0	2.0	3.0	5.0	1.0	1.0	2.0	3.0	15.0	-21.0	MIN	Minnesota Timberwolves	2022-02-25	2021	0	10245480.0	10245480	137098327	137098327	4125163242	4125163242	0.074731	0.002484	0.033235

df.columns

Index(['PLAYER_ID', 'PLAYER_NAME', 'START_POSITION', 'MIN', 'FGM', 'FGA',
       'FG_PCT', 'FG3M', 'FG3A', 'FG3_PCT', 'FTM', 'FTA', 'FT_PCT', 'OREB',
       'DREB', 'REB', 'AST', 'STL', 'BLK', 'TO', 'PF', 'PTS', 'PLUS_MINUS',
       'TEAM_ABBREVIATION', 'TEAM_NAME', 'GAME_DATE_EST', 'SEASON_START',
       'WON_GAME', 'SALARY', 'INFLATION_ADJ_SALARY', 'TEAM_PAYROLL',
       'INFLATION_ADJ_TEAM_PAYROLL', 'LEAGUE_PAYROLL',
       'INFLATION_ADJ_LEAGUE_PAYROLL', 'TEAM_IMPORTANCE', 'LEAGUE_IMPORTANCE',
       'TEAM_MARKET_SIZE'],
      dtype='object')

df = df.loc[df['MIN'] != '0:00']

Scatter Plot Matrix

We want to create a matrix of scatter plots and actually visualize the correlations between each numerical variable.

avg_df = df.groupby(['SEASON_START', 'PLAYER_ID']).mean()
avg_df.head()

		FGM	FGA	FG_PCT	FG3M	FG3A	FG3_PCT	FTM	FTA	FT_PCT	OREB	DREB	REB	AST	STL	BLK	TO	PF	PTS	PLUS_MINUS	WON_GAME	SALARY	INFLATION_ADJ_SALARY	TEAM_PAYROLL	INFLATION_ADJ_TEAM_PAYROLL	LEAGUE_PAYROLL	INFLATION_ADJ_LEAGUE_PAYROLL	TEAM_IMPORTANCE	LEAGUE_IMPORTANCE	TEAM_MARKET_SIZE
SEASON_START	PLAYER_ID
2003	15	1.440000	3.840000	0.343820	0.860000	2.440000	0.309060	0.280000	0.320000	0.143340	0.200000	1.280000	1.480000	0.520000	0.320000	0.100000	0.580000	0.860000	4.020000	-0.300000	0.440000	2500000.0	3697550.0	55401430.0	81939823.0	1.672617e+09	2.473835e+09	0.045125	0.001495	0.033123
	43	1.133333	2.800000	0.311267	0.533333	1.133333	0.283333	0.333333	0.333333	0.200000	0.200000	0.666667	0.866667	0.933333	0.400000	0.133333	0.400000	0.400000	3.133333	-3.466667	0.200000	1070000.0	1582551.0	45682902.0	67565925.0	1.672617e+09	2.473835e+09	0.023422	0.000640	0.027312
	56	5.288462	11.634615	0.450538	0.663462	2.125000	0.250154	1.932692	2.692308	0.528808	0.875000	3.105769	3.980769	5.432692	1.144231	0.221154	1.721154	2.201923	13.173077	3.076923	0.663462	4917000.0	7272342.0	64123085.0	94839328.0	1.672617e+09	2.473835e+09	0.076681	0.002940	0.038337
	57	3.968085	8.819149	0.438585	0.659574	1.861702	0.246202	2.010638	2.340426	0.608915	0.914894	3.372340	4.287234	4.191489	1.840426	0.489362	1.882979	2.351064	10.606383	3.882979	0.659574	6900000.0	10205238.0	68372826.0	101124779.0	1.672617e+09	2.473835e+09	0.100917	0.004125	0.040878
	72	2.395833	5.479167	0.394979	0.020833	0.083333	0.020833	0.729167	1.000000	0.288187	0.395833	1.312500	1.708333	2.625000	0.562500	0.104167	1.145833	1.583333	5.541667	1.312500	0.750000	1070000.0	1582551.0	58156236.0	86014236.0	1.672617e+09	2.473835e+09	0.018399	0.000640	0.034770

sn.pairplot(avg_df)

<seaborn.axisgrid.PairGrid at 0x2070213a8b0>

League Importance of Players

How does the lowest, highest, average league importance (i.e., the proportion of league payroll that a player's salary takes up) change over time?

minweights = df.groupby(['SEASON_START'])["LEAGUE_IMPORTANCE"].min()
maxweights = df.groupby(['SEASON_START'])["LEAGUE_IMPORTANCE"].max()
avgweights = df.groupby(['SEASON_START'])["LEAGUE_IMPORTANCE"].mean()

years = df['SEASON_START'].unique()

plt.figure(figsize=(20,15))
plt.scatter(years, minweights)
plt.scatter(years, maxweights)
plt.scatter(years, avgweights)
plt.locator_params('x', nbins = 20)
plt.xlabel('Year')
plt.ylabel('League Importance')
plt.title('Minimum, Average, and Maximum Player Salary Weights on League Payroll')

Text(0.5, 1.0, 'Minimum, Average, and Maximum Player Salary Weights on League Payroll')